© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13703
    Prev
    Next
    INDEX
    Explanations

    quotes followed by modal or auxiliary verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     a
    -1.74
     even
    -1.65
     an
    -1.65
     melap
    -1.63
     بودند
    -1.63
     tersenyum
    -1.61
     ANCHE
    -1.48
     faſt
    -1.48
     anything
    -1.45
     menyes
    -1.45
    POSITIVE LOGITS
     will
    2.81
     are
    2.58
     is
    2.55
     was
    2.45
     can
    2.30
     would
    2.19
     should
    2.09
     has
    2.03
     could
    2.00
     must
    1.95
    Activations Density 0.029%

    No Known Activations